If you have a between subject design, use Univariate - within subject, use Repeated Measures, and mixed, Repeated Measures.

When with 2 (or more) repeated measures variables, SPSS holds the initial variable(s) constant and varies the later one(s).  

You should be able to add, subtract, multiply and divide as well as compute SDs, MEANs and  SUMs of variables. 

• File: New: Syntax (Normally used if your are an advanced programmer, or are simply copying in syntax  from another soure, for example your wise and thoughtful instructor.)
• While in a Transform or Statististical function, click on Paste (This is most common for beginning programmers because you often do one transform, or statistical function, to learn how the relevant syntax should be written.)

• When you have more than one variable--Analyze : Descriptive Statistics:Crosstabs: Remember to click on Statistics:Chi-Square 
• When you have one variable--Analyze: Nonparametric Tests: Chi-Square

First you would Compute the new variables, lets say  msum and fsum, and make them equal to 0 (This pre-compute is necessary for the sum below to work, but isn't always necessary ).   Having done this we can now Do Repeat.

do repeat x = att1 to att13/y = obsex1 to obsex13. 
if y = 1
fsum = fsum + x.
if y = 2
msum = msum + x.
end repeat.
execute. 

DON'T MISS ANY PERIODS!!!!

The following would list height and weights. 

List Variables=height weight. 
 

numeric qa1 to qa200 qb1 to qb50 (f1.0) qc1 to qc10 (f3.1).
execute.

The preceding creates three strings of variables--qa1 to qa200 and qb1 to qb50, each one didget long with no decimal places and qc1 to qc10 with three didgets and one decimal place.  The "execute" runs the preceding line. 

Note!! This command only works if you have at least one variable (even if it's empty of data) in your dataset.  I often simply creat a "junk" variable, run the numeric syntax, and then delete the junk variable.

Click on Options and Means and SD if want to show means and SD, then click on Continue.

If using different variable: move Variable to right, under Output Variable: Name: put new variable name. Click change. Click on Old and New values, put original value in Value under Old Value . Put new value in Value under New Value. Click Add. Repeat steps until you have all values recoded. Click Continue. Click okay.

If using same variable: follow the steps above except ignore Output: Name directions since there is no new variable. 

(using Compute )– 

Symbols: And  = &
                Or   = |
You must repeat the variable name after every Boolean statement. 

 You should be able to use the following symbols:

Also under Recode: follow same directions.

or 

Split File

(Split File is the preferred method!!  Use it on your proficiency test if you are asked to do something like this....)

Temporary. 
Select If (gender = 2). 
CORRELATIONS 
    /VARIABLES=weight height
    /PRINT=TWOTAIL NOSIG 
    /MISSING=PAIRWISE. 

Other ways to do the above analysis would be to Compute a new variable that only exists If gender = 2 and then run your analysis on the new variable.  You could also go to Data: Select Cases: If. 

Another cute option is to go to Data: Split File: and then split the file based on gender.  Once this is done all subsequent tests will be done separately on each of the splitting variables, until you turn off the Split File.  One Pearson will generate one for "males" and one for "females," for example.  We recommend that you use Split File rather than Temporary Select if because if the Temporary part fails, then some of yourdata is thrown out. 

The syntax for splitting a file is as follows (again, notice the placement of periods):

(The following goes prior to your analyses:)
SORT CASES by variable name
SPLIT FILE
   LAYERED BY variable name

(The following goes at the end of your analyses)
SPLIT FILE 
 OFF. 

If using independent samples: Under group Variable insert name of variable you want to compare them on.

You can also use syntax:

value labels gender 1 'Female' 2 'Male'.
execute.

Here an existing variable (gender) gets appropriate labels associated with the numeric values in the data set.  The "execute" simply makes the preceding line "execute," as it were. 

Warning: You might notice that when you open up your SPSS file, that under a variable instead of having a number (e.g, 1)  you will have a word (e.g, Male).  This simply means that the computer is set up to show value labels.  To fix this, simply got to View : and click off value labels. 

Syntax would read:

variable labels  x 'Whatever You Want'.
execute.

Here "x" is an existing variable name--the rest is self-explanatory.  You can do multiple variables, if you'd like, just put / between each and it's best to not to go to the next line unless you've just /'d.